OpenCV And Depth Camera Spots Weeds

January 31, 2021 by 25 Comments

Using vision technology to identify weeds in agriculture is an area of active development, and a team of researchers recently shared their method of using a combination of machine vision plus depth information to identify and map weeds with the help of OpenCV, the open-source computer vision library. Agriculture is how people get fed, and improving weed management is one of its most important challenges.

Many current efforts at weed detection and classification use fancy (and expensive) multispectral cameras, but PhenoCV-WeedCam relies primarily on an OAK-D stereo depth camera. The system is still being developed, but is somewhat further along than a proof of concept. The portable setups use a Raspberry Pi, stereo camera unit, power banks, an Android tablet for interfacing, and currently require an obedient human to move and point them.

It’s an interesting peek at the kind of hands-on work that goes into data gathering for development. Armed with loads of field data from many different environments, the system can use the data to identify grasses, broad leaf plants, and soil in every image. This alone is useful, but depth information also allows the system to estimate overall plant density as well as try to determine the growth center of any particular plant. Knowing that a weed is present is one thing, but to eliminate it with precision — for example with a laser or mini weed whacker on a robot arm — knowing where the weed is actually growing from is an important detail.

PhenoCV-WeedCam (GitHub repository) is not yet capable of real-time analysis, but the results are promising and that’s the next step. The system currently must be carried by people, but could ultimately be attached to a robotic platform made specifically to traverse fields.

Robotic Fish Swarm Together Using Cameras And LEDs

January 18, 2021 by 4 Comments

Robotics has advanced in leaps and bounds over the past few decades, but in terms of decentralized coordination in robot swarms, they far behind biological swarms. Researchers from Harvard University’s Weiss Institute are working to close the gap, and have developed Blueswarm, a school of robotic fish that can exhibit swarm behavior without external centralized control.

In real fish schools, the movement of an individual fish depends on those around it. To allow each robotic fish to estimate the position of its neighbors, they are equipped with a set of 3 blue LEDs, and a camera on each side of the body. Four oscillating fins, inspired by reef fish, provide 3D control. The actuator for the fins is simply a pivoting magnet inside a coil being fed an alternating current. The onboard computer of each fish is a Raspberry Pi W, and the cameras are Raspberry Pi Camera modules with wide-angle lenses. Using the position information calculated from the cameras, the school can coordinate its movements to spread out, group together, swim in a circle, or find an object and then converge on it. The full academic article is available for free if you are interested in the details.

Communication with light is dependent on the clarity of the medium it’s traveling through, in this case, water — and conditions can quickly become a limiting factor. Submarines have faced the same challenge for a long time. Two current alternative solutions are ELF radio and sound, which are both covered in [Lewin Day]’s excellent article on underwater communications.

Continue reading “Robotic Fish Swarm Together Using Cameras And LEDs” →

The Game Boy Camera, Or: How I Learned To Stop Worrying And Love The Pixels

October 26, 2020 by 11 Comments

Never underestimate the power of nostalgia. In an age when there are more megapixels stuffed in the sensor of a smartphone camera than the average computer display can even represent, why would jagged images from a 20-year-old grayscale camera with pixels numbering in the thousands still grab attention? Maybe what’s old is new again, and the coolness factor of novelty is something that can’t be quantified.

The surprise I had last Monday when I saw my Twitter notifications is maybe only second to the feeling I had when I was invited to become a Hackaday contributor. I’d made a very simple web app which mimics a Game Boy Camera using the camera from your phone or desktop, and it got picked up by people so much that I’m amazed my web host is still holding. Let’s look at why something seemingly so simple gained so much traction.

Continue reading “The Game Boy Camera, Or: How I Learned To Stop Worrying And Love The Pixels” →

The Floppy Disk As A Portable Music Format

September 7, 2020 by 27 Comments

We remember the floppy disk as the storage medium most of us used two decades or more ago, limited in capacity and susceptible to data loss. It found its way into a few unexpected uses such as Sony’s Mavica line of digital cameras, but outside those who maintain and use older equipment it’s now ancient history.

Seemingly not for [Terence Eden] though, who has made a portable audio player that uses a floppy disk as its storage medium. It came about with the realization that half an hour of extremely compressed audio could be squeezed onto a standard 3.5 inch floppy, and then that the Beatles’ A Hard Day’s Night album comes in at only a shade over that time. With some nifty manipulation of the compression command line and the judicious removal of some unnecessary metadata, the album can fit on a floppy in equivalent quality to the AM radio fans would have heard it over back in 1964.

The player would have been a major undertaking when the floppy was king, but in 2020 it’s simply a USB floppy drive, a Raspberry Pi, and a battery pack. He’s given us the full instructions, and no doubt a more permanent version could be built with a 3D-printed case.

We’re fascinated by the recent trend of storing audio on floppy disks, but despite the hipster vibe, we doubt  the idea will catch on. It’s not the first floppy-based player we’ve seen, but the previous one was more of a fake player.

Why Are Digital Cameras Still Boring?

July 1, 2020 by 102 Comments

In the matter of technological advancement, we are as a species, mostly insatiable. The latest toy, the fastest silicon, the largest storage, the list goes on. Take digital cameras as an example, what was your first one? Mine was a Casio QV200 in about 1997, I still have it somewhere though I can’t immediately lay my hands on it, and it could hold a what was for its time a whopping 64 VGA-resolution pictures in its 4Mb of onboard memory.

The QV200 showing off its VGA capabilities. It's March 1998, and this is a brand-new PlayStation that I'm about to install a mod chip inside.
The QV200 showing off its VGA photography capabilities. It’s March 1998, and this is a brand-new PlayStation that I’m about to install a mod chip inside.

It’s a shock to realise that nearly a quarter century has passed since then, and its fixed-focus 640×480 camera module with a UV-sensitive CMOS sensor that gave everything a slight blue tint would not even grace the cheapest of feature phones in 2020. Every aspect of a digital camera has improved beyond measure since the first models in the 1980s and early 1990s that started to resemble what we’d know today as a standalone digital camera, they have near-limitless storage, excellent lenses, huge and faithfully-reproducing sensors, and broadcast-quality video capability.

But how playful have camera manufacturers been with the form factor? We see reporters in sci-fi movies toting cameras that look nothing like their film-based ancestors. What do our real-life digital cameras have on offer as far as creative body design goes?

Continue reading “Why Are Digital Cameras Still Boring?” →

This Camera Captures Piezo Inkjet Micro-Drops For DIY Microfluidics

April 15, 2020 by 6 Comments

In microfluidics, there are “drop on demand” instruments to precisely deposit extremely small volumes (pico- or nano-liters) of fluid. These devices are prohibitively expensive, so [Kyle] set out to design a system using hobbyist-level parts for under $1000. As part of this, he has a fascinating use case for a specialized camera: capturing the formation and shape of a micro-drop as it is made.

There are so many different parts to this effort that it’s all worth a read, but the two big design elements come down to:

  1. Making the microdrop using a piezo element
  2. Ensuring the drop is made correctly, and visually troubleshooting
Working prototype. The piezo tube is inside the blue piece at the top. The camera is to the right, and the LED strobe is on the left.

It’s one thing to make an inkjet element in a printer work, but it’s quite another to make a piezoelectric element dispense arbitrary liquids in a controlled, repeatable, and predictable way. Because piezoelectric elements force liquid out with a mechanical motion, different liquids require different drive signals and that kind of experimentation requires a way to see what is going on, hence the need for a drop observation camera.

[Kyle] ended up taking the lens assembly from a cheap USB microscope and mating it to his Korukesu C1 USB Camera with a 3D printed assembly. Another 3D printed enclosure doubles as a lightbox, holding the piezo tube in the center with the LED strobe and camera on opposite sides. The whole assembly had a few false starts, but in the end [Kyle] seems pretty happy with his results. The device is briefly described at a high level here. There are some rough edges, but it’s a working system.

Inkjet technology has been around for a long time (you can see a thirty-plus year old inkjet printer in action here) but it’s worth mentioning that not all inkjet heads are alike. Most inkjet printer heads operate thermally, which means a flash of heat vaporizes some ink to expel a micro-drop. These heads aren’t very suitable for microfluidics because not only do they rely on vaporizing the liquid, but they also don’t work well with anything other than the ink they’re designed for. Piezoelectric print heads are less common, but are more suited to the kind of work [Kyle] is doing.

Recording Video In The Era Of CRTs: The Video Camera Tube

February 27, 2020 by 45 Comments

We have all watched videos of concerts and events dating back to the 1950s, but probably never really wondered how this was done. After all, recording moving images on film had been done since the late 19th century. Surely this is how it continued to be done until the invention of CCD image sensors in the 1980s? Nope.

Although film was still commonly used into the 1980s, with movies and even entire television series such as Star Trek: The Next Generation being recorded on film, the main weakness of film is the need to move the physical film around. Imagine the live video feed from the Moon in 1969 if only film-based video recorders had been a thing.

Let’s look at the video camera tube: the almost forgotten technology that enabled the broadcasting industry. Continue reading “Recording Video In The Era Of CRTs: The Video Camera Tube” →